Messina et al. J Transl Med (2020) 18:233 https://doi.org/10.1186/s12967-020-02405-w Journal of Translational Medicine RESEARCH Open Access COVID-19: viral–host interactome analyzed by network based-approach model to study pathogenesis of SARS-CoV-2 infection Francesco Messina1†, Emanuela Giombini1†, Chiara Agrati1, Francesco Vairo1, Tommaso Ascoli Bartoli1, Samir Al Moghazi1, Mauro Piacentini1,2, Franco Locatelli3, Gary Kobinger4, Markus Maeurer5,6, Alimuddin Zumla7,8, Maria R. Capobianchi1* , Francesco Nicola Lauria1†, Giuseppe Ippolito1† and COVID 19 INMI Network Medicine for IDs Study Group Abstract Background: Epidemiological, virological and pathogenetic characteristics of SARS-CoV-2 infection are under evaluation. A better understanding of the pathophysiology associated with COVID-19 is crucial to improve treatment modalities and to develop efective prevention strategies. Transcriptomic and proteomic data on the host response against SARS-CoV-2 still have anecdotic character; currently available data from other coronavirus infections are there- fore a key source of information. Methods: We investigated selected molecular aspects of three human coronavirus (HCoV) infections, namely SARS- CoV, MERS-CoV and HCoV-229E, through a network based-approach. A functional analysis of HCoV–host interactome was carried out in order to provide a theoretic host–pathogen interaction model for HCoV infections and in order to translate the results in prediction for SARS-CoV-2 pathogenesis. The 3D model of S-glycoprotein of SARS-CoV-2 was compared to the structure of the corresponding SARS-CoV, HCoV-229E and MERS-CoV S-glycoprotein. SARS-CoV, MERS-CoV, HCoV-229E and the host interactome were inferred through published protein–protein interactions (PPI) as well as gene co-expression, triggered by HCoV S-glycoprotein in host cells. Results: Although the amino acid sequences of the S-glycoprotein were found to be diferent between the various HCoV, the structures showed high similarity, but the best 3D structural overlap shared by SARS-CoV and SARS-CoV-2, consistent with the shared ACE2 predicted receptor. The host interactome, linked to the S-glycoprotein of SARS-CoV and MERS-CoV, mainly highlighted innate immunity pathway components, such as Toll Like receptors, cytokines and chemokines. Conclusions: In this paper, we developed a network-based model with the aim to defne molecular aspects of pathogenic phenotypes in HCoV infections. The resulting pattern may facilitate the process of structure-guided phar- maceutical and diagnostic research with the prospect to identify potential new biological targets. Keywords: Coronavirus infection, Virus–host interactome, Spike glycoprotein Background *Correspondence: [email protected] In December 2019, a novel coronavirus (SARS-CoV-2) †Francesco Messina, Emanuela Giombini, Francesco Nicola Lauria and Giuseppe Ippolito contributed equally to this work was frst identifed as a zoonotic pathogen of humans 1 National Institute for Infectious Diseases “Lazzaro Spallanzani” IRCCS, in Wuhan, China, causing a respiratory infection Rome, Italy with associated bilateral interstitial pneumonia. Te Full list of author information is available at the end of the article © The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/publi cdoma in/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Messina et al. J Transl Med (2020) 18:233 Page 2 of 10 disease caused by SARS-CoV-2 was named by the We investigated here biologically and clinically relevant World Health Organization as COVID-19 and it has molecular targets of three human coronaviruses (HCoV) been classifed as a global pandemic since it has spread infections using a network based approach. A functional rapidly to all continents. As of May 20, 2020, there have analysis of HCoV–host interactome was carried out in been 4.889.287 confrmed COVID-19 cases worldwide order to provide a theoretic host–pathogen interaction with 322.457 deaths reported to the WHO [1]. Whilst model for HCoV infections, and to predict viable mod- clinical and epidemiological data on COVID-19 have els for SARS-CoV-2 pathogenesis. Tree HCoV caus- become readily available, information on the pathogen- ing respiratory diseases were used as the model targets, esis of the SARS-CoV-2 infection has not been forth- namely: SARS-CoV, that shares with SARS-CoV-2 a coming [2]. Te transcriptomic and proteomic data on strong genetic similarity, including MERS-CoV, and host response against SARS-CoV-2 is scanty and not HCoV-229E. efective therapeutics and vaccines for COVID-19 are available yet. Methods Coronaviruses (CoVs) typically afect the respiratory Comparative reconstruction of S‑glycoprotein in HCoVs tract of mammals, including humans, and lead to mild Te reconstruction of virus–host interactome was car- to severe respiratory tract infections [3]. Many emerg- ried out using the RWR algorithm to explore the human ing HCoV infections have spilled-over from animal res- PPI network and the multilayer PPI platform enriched ervoirs, such as HCoV-OC43 and HCoV-229E which with gene expression data sets. 259 sequences of CoVs, cause mild diseases such as common colds [4, 5]. During infecting diferent animal hosts (Additional fle 1: the past 2 decades, two highly pathogenic HCoVs, severe Table S1), were downloaded by GSAID and NCBI data- acute respiratory syndrome coronavirus (SARS-CoV) and base in order to evaluate the variability in the S gene. Middle East respiratory syndrome coronavirus (MERS- SARS-CoV, HCoV-229E and MERS-CoV and other CoV CoV), have led to global epidemics with high morbid- full genome sequence groups were aligned with MAFFT ity and mortality [6]. In this period, a large amount of [14], synonymous and non-synonymous mutations, and experimental data associated with the two infections has amino acid similarity were calculated using the SSE pro- allowed to better understand molecular mechanism(s) of gram with a sliding windows of 250 nucleotides and a coronavirus infection, and enhance pathways for devel- pass of 25 nu [15]. A homology model was built for the oping new drugs, diagnostics and vaccines and identi- amino acid sequences of the S-glycoprotein, derived from fcation of host factors stimulating (proviral factors) or the full genome sequence obtained at “SARS-CoV-2/ restricting (antiviral factors) infection remains poorly INMI1/human/2020/ITA” (MT066156.1). Te Swiss pdb understood [7]. Structures of many proteins of SARS- server was used to construct three-dimensional mod- CoV and MERS-CoV, and biological interactions with els for the S-glycoprotein of SARS-CoV-2 [16]. Among other viral and host proteins have been widely explored; proteins with a 3D structure, the best match with the through experimental testing of small molecule inhibi- “SARS-CoV-2/INMI1/human/2020/ITA” sequence was tors with anti-viral efects [8, 9]. ACE2, expressed in type the 6VSB.1, that was evaluated considering the iden- 2 alveolar cells in the lung, has been identifed as recep- tity of two amino acid sequences and the QMEN value tor of SARS-CoV and SARS-CoV-2, while dipeptidyl included in Swiss pdb server. Te model of a single chain peptidase DPP4 was identifed as the specifc receptor for was overlapped with the three-dimensional structure MERS-CoV [10, 11]. of S-glycoprotein single chain belonging to SARS-CoV Te investigation of structural genomics and inter- (5WRG), HCoV-229E (6U7H.1) and MERS-CoV (5X59), actomics of SARS-CoV-2 can be implemented through using Chimera 1.14 [17]. In order to better evaluate the systematical mapping of protein–protein interactions conservation of the sequence in each site, all sequences (PPI) between SARS-CoV-2 and human host, and an were aligned with MAFFT and the topology of all struc- integrated bioinformatics approach [12, 13]. Structural tures were compared. Te detailed description of the analysis of specifc SARS-CoV-2 proteins, in particular reconstruction of S-glycoprotein structure is reported in Spike glycoproteins (S-glycoproteins), and their interac- Additional fle 2. tions with human proteins, can guide the identifcation of the putative functional sites and help to better defne the PPI and gene co‑expression network pathologic phenotype of the infection. Tis functional Network analysis, based on protein–protein interactions interaction analysis between the host and other HCoVs, and gene expression data, was performed in order to view combined with an evolutionary sequence analysis of all possible virus–host protein interactions during the SARS-CoV-2, can be used to guide new treatment and HCoV infections. Since the SARS-CoV-2 genome exhibits prevention interventions. substantial similarity to the SARS-CoV genome [18] and Messina et al. J Transl Med (2020) 18:233 Page 3 of 10 subsequently also the proteome [19], we hypothesized sequence obtained at Laboratory of Virology, National that several molecular interactions that were observed Institute for Infectious Diseases “L. Spallanzani” IRCCS, in the SARS-CoV interactome will be preserved in the using Swiss pdb viewer server (Additional fle 2: Figure SARS-CoV-2 interactome.